[Paper Review] Relationship between Cross-Polarization Discrimination (XPD) and Spatial Correlation in Indoor Small-Cell MIMO Systems

Yeon-Geun Lim, Yae Jee Cho+3 more|arXiv (Cornell University)|Jul 2, 2017

Advanced MIMO Systems Optimization12 references3 citations

TL;DR

This paper proposes a correlated MIMO channel model that links cross-polarization discrimination (XPD) to spatial correlation in indoor small-cell systems using dual-polarization antennas. By leveraging 3D ray-tracing simulations, it demonstrates that higher XPD reduces spatial correlation, enabling equivalent antenna spacing estimates; notably, XPD values of 20 dB and 30 dB yield negligible throughput differences, suggesting cost-effective design trade-offs for compact MIMO arrays.

ABSTRACT

In this letter, we present a correlated channel model for a dual-polarization antenna to omnidirectional antennas in indoor small-cell multiple-input multiple-output (MIMO) systems. In an indoor environment, we confirm that the cross-polarization discrimination (XPD) in the direction of angle-of-departure can be represented as the spatial correlation of the MIMO channel. We also evaluate a dual-polarization antenna-based MIMO channel model and a spatially correlated channel model using a three-dimensional (3D) ray-tracing simulator. Furthermore, we provide the equivalent distance between adjacent antennas according to the XPD, providing insights into designing a dual-polarization antenna and its arrays.

Motivation & Objective

• To model the relationship between XPD and spatial correlation in dual-polarization MIMO systems for indoor small cells.
• To evaluate the impact of XPD on MIMO channel capacity using a 3D ray-tracing simulator.
• To derive equivalent antenna spacing based on XPD to guide compact dual-polarization antenna array design.
• To provide design insights that balance performance, cost, and complexity in small BSs.

Proposed method

• Develops an effective channel model using a polarization transfer matrix with copolarized and cross-polarized propagation gains.
• Uses 3D ray-tracing to simulate realistic indoor channel behavior, including multipath and polarization effects.
• Applies isotropic and Laplacian angle-of-departure (AoD) distributions to model spatial correlation in the effective channel matrix.
• Derives spatial correlation coefficients via expectation over random phases and AoD distributions.
• Calculates equivalent antenna separation distances by matching spatial correlation from dual-polarization channels to omnidirectional arrays.
• Validated through ergodic throughput comparisons using cumulative distribution functions (CDFs) from simulation results.

Experimental results

Research questions

• RQ1 How does XPD in the angle-of-departure direction affect spatial correlation in dual-polarization MIMO channels?
• RQ2 What is the equivalent antenna spacing between omnidirectional antennas that matches the spatial correlation of a dual-polarization antenna for a given XPD?
• RQ3 How does XPD influence MIMO system throughput in indoor small-cell environments?
• RQ4 What is the performance trade-off between high XPD and system complexity in compact MIMO arrays?
• RQ5 Can a 3D ray-tracing-based model accurately reflect the relationship between XPD and spatial correlation?

Key findings

• XPD has a strong inverse relationship with spatial correlation: as XPD increases, spatial correlation decreases significantly.
• At XPD = 20 dB, the spatial correlation coefficient drops to 0.1980, and at XPD = 30 dB, it further reduces to 0.0632.
• The equivalent antenna spacing for Laplacian AoD distribution increases from 0.100λ at 3 dB XPD to 0.850λ at 30 dB XPD.
• Throughput differences between XPD = 20 dB and 30 dB are negligible, indicating diminishing returns beyond 20 dB.
• The equivalent spacing for isotropic AoD distribution increases from 0.076λ to 0.364λ across the same XPD range.
• The proposed effective correlation matrix enables accurate modeling and design of compact dual-polarization antenna arrays with minimal performance degradation.